ENHANCING HEART HEALTH: PERFORMANCE ANALYSIS & COMPARISON OF SUPERVISED MACHINE LEARNING ALGORITHMS FOR CARDIOVASCULAR DISEASE PREDICTION

Abstract

Heart disease, commonly known as cardiovascular disease, has become a leading cause of death globally. It includes numerous disorders that affect the heart and has been a major cause of death around the world in the previous few decades [1], about 26 million people effects every year. The prediction and prevention of heart failure is a challenge for cardiologists and cardio-surgeons.

 The modern lifestyle, poor diet, lack of exercise, and high stress and depression level also increase the rate of cardiovascular disease. Early detection of cardiovascular disease signs and consistent medical monitoring can help decrease in the number of patients and mortality, but however, with limited medical resources and specialist consultants, it is difficult to continuously observe the patient and provide consultation.

 The healthcare industry holds a substantial amount of data, making machine learning algorithms essential for accurately predicting heart diseases and facilitating informed decision- making. Recent studies have explored the integration of these approaches to create hybrid machine learning algorithms. In this research, some of the data pre-processing techniques, such as eliminating noisy data, eliminating missing data, filling in default values where applicable, and splitting attributes into categories for predictions and decision making across different levels, this project suggests the development of a predictive model to determine the likelihood of individuals having a heart disease, aiming to offer both awareness and diagnostic insights.

 The accuracy of several techniques, such as Support Vector Machine, Logistic Regression, Random Forest Classifier, Naive Bayes Classifier, and K-Nearest Neighbour, are compared in order to achieve this goal. The goal of this comparison is to determine which model predicts cardiovascular disease the most accurately.